Advances in Deep Learning for Air Pollution and Climate Science Research

Dear Colleagues,

This Special Issue, Advances in Deep Learning for Air Pollution and Climate Science Research, invites manuscripts that present methodologically novel AI approaches for air quality and climate science. We seek studies that push the AI frontier for air quality and climate science by introducing new learning paradigms and model architectures, and by obtaining scientific insights into atmospheric processes from machine learning models. This aligns with the scope and reproducibility expectations of the MDPI journal AI.

We welcome advances in spatiotemporal learning for air pollutant fields. This includes transformer-style sequence models for long-horizon prediction and multi-resolution dynamics. We also welcome graph-based learning that represents transport and source–receptor structure. Physics-informed and hybrid AI–physics methods are encouraged. Generative approaches are welcome for the downscaling, super-resolution, scenario simulation, and imputation of missing data. We encourage the application and fine-tuning of foundation models for air quality and climate tasks. Remote sensing retrieval and satellite-informed learning are within the scope of the Special Issue. We also welcome machine-learning-based data assimilation, including inverse modeling for emissions and source attribution. AI emulators that accelerate chemical transport models (CTM) and Earth system models (ESM) are covered. Explainable AI and uncertainty quantification are encouraged to support process understanding and calibrated prediction intervals.

Topics of interest include, but are not limited to, the following:

• Spatiotemporal AI for air pollutant estimation and forecasting.
• Transformer-style models for long-horizon and multi-resolution prediction.
• Graph learning for transport and source–receptor relationships.
• Physics-informed and hybrid AI–physics modeling.
• Foundation model fine-tuning for air quality and climate science applications.
• Remote sensing retrieval and satellite-informed air quality methods.
• Data assimilation and inverse modeling for emissions and source attribution.
• AI emulators for chemical transport and Earth system models.
• Generative AI for downscaling, imputation, and scenario simulation.
• Explainable AI and uncertainty quantification for process understanding and calibrated prediction intervals.

Dr. Alqamah Sayeed
Dr. Ahmed Khan Salman
Guest Editors

Manuscript Submission Information

Manuscripts should be submitted online at www.mdpi.com by registering and logging in to this website. Once you are registered, click here to go to the submission form. Manuscripts can be submitted until the deadline. All submissions that pass pre-check are peer-reviewed. Accepted papers will be published continuously in the journal (as soon as accepted) and will be listed together on the special issue website. Research articles, review articles as well as short communications are invited. For planned papers, a title and short abstract (about 250 words) can be sent to the Editorial Office for assessment.

Submitted manuscripts should not have been published previously, nor be under consideration for publication elsewhere (except conference proceedings papers). All manuscripts are thoroughly refereed through a single-blind peer-review process. A guide for authors and other relevant information for submission of manuscripts is available on the Instructions for Authors page. AI is an international peer-reviewed open access monthly journal published by MDPI.

Please visit the Instructions for Authors page before submitting a manuscript. The Article Processing Charge (APC) for publication in this open access journal is 1800 CHF (Swiss Francs). Submitted papers should be well formatted and use good English. Authors may use MDPI's English editing service prior to publication or during author revisions.

• air pollution estimation and forecasting
• climate variability and extremes
• satellite retrievals for air quality
• wildfire smoke detection and exposure mapping
• bias correction for CTM and ESM outputs
• CTM and ESM emulation
• downscaling and super-resolution
• data assimilation and inverse modeling
• foundation model fine-tuning
• explainable AI and uncertainty quantification